基于Hadoop的电子商务推荐系统的设计与实现

Design and implementation of recommendation system for E-commerce on Hadoop

为了解决大数据应用背景下大型电子商务系统所面临的信息过载问题,研究了基于Hadoop构建分布式电子商务推荐系统的方案。采用基于MapReduce模型实现的算法具有较高的伸缩性和性能,能高效地进行离线数据分析。为了克服单一推荐技术的不足,设计了融合多种互补性推荐技术的混合推荐模型。实验结果表明,基于Hadoop平台实现的推荐系统具有较好的伸缩性和性能。

To solve the information overload problem of large scale E-commerce systems in the big data era, a solution based on Hadoop is proposed, aiming at building a distributed recommendation system. Data analysis algorithms based on MapReduce programming model have high scalability and good performance. To overcome the limit of single recommendation technology, a hybrid model is adopted, which combines several complementary methods. Empirical studies show that the recommendation system on Hadoop has good scalability and efficiency.